Available from Version: 2.0
This tool determines normalised expression levels of sRNAs matching known miRNAs in miRBase. miRProf can also quickly and easily compare miRNA expression levels across multiple samples.
miRProf requires a set of sRNA samples and a genome file for input. The paths to the files representing the samples and the genome are entered at the top of the main dialog box. In addition, miRProf can be configured in a number of ways, which are described in the following paragraphs.
miRProf filters sRNA sequences in each sample before expression levels are computed. The user can specify criteria such as minimum sRNA size, maximum sRNA size and minimum abundance. In addition, all low-complexity sRNAs (sequences, which contain less than 3 distinct nucleotides) and sequences containing non-canonical nucleotides (e.g. ‘N’ for unknown nucleotide) are discarded. Finally, any sRNAs not aligning to the user-specified genome are discarded. Genome filtering is mandatory in order to provide a frame of reference for comparing sRNAs across multiple samples.
Filtered reads from each sample are then aligned to a (sub)set of miRBase mature sequences. The user has the option to partion mirBase mature sequence based on whether they originate from plant, animal or virus miRNA precursors. In addition, the user can optionally choose to include all sets for alignment. miRBase is constantly updated with new miRNAs reported from the field. Therefore, the user can select any version of miRBase for their specific experiments. The data download and management is automatically handled within the tool. All the user need do is type in the version they which to use in the parameters dialog box.
The sRNA to mature miRNA matching process can be controlled using a few parameters:
- Overhangs allowed
When aligning sRNAs to mature sequences the aligning tool will fail to report a match if the short read (the sRNA) overhangs the long read (the miRNA), to circumvent this problem miRNAs are padded with “XXX” at the beginning and end of the sequence. sRNAs which overhang the actual mature sequence will then be treated as having a number of mismatches equal to the number of nucleotides, which overhang the miRNA. This behaviour occurs by default, but the user can switch this off by deselecting the “overhangs allowed” check box.
- Mismatches allowed
The number of allowed mismatches is controlled using the “Mismatches allowed” text box. The default value here is 0, implying the sRNA must exact match somewhere within the mature sequence. The maximum number of mismatches allowed is capped at 3 to prevent huge numbers of meaningless results being recorded.
- Only keep best match
Even with 1, 2, or 3 mismatches allowed many hits can be returned, these can be minimised by checking the “only keep best match” box, which instructs miRProf to discard all hits to a given miRNA that have less than the best number of mismatches (a smaller number of mismatches is considered “better” than a larger number).
The last part of miRProf configuration involves determining how the sRNAs that match known miRNAs should be grouped. The user has these options:
- Group organisms
If checked, miRProf does not consider the organism from which the miRNA was found. Therefore, users who select this check box will see only a single organism being returned: (all-combined).
- Group variants
If checked, matches to different variants of a single miRNA precursor family are combined into one, such as: arm, and copy number.
- Group mature and star
If checked, matches to mature and star sequences of the derived from the same miRNA precursor are grouped into one.
- Group mismatches
If checked, matches to the same miRNA are combined into a group regardless of the number of mismatches.
All miRProf configuration parameters can be saved and loaded from file by using the toolbar buttons at the top of parameter browser dialog. This functionality can help the user maintain consistency between separate runs and can be used as part of an experiment log book by automatically documenting the experimental setup.
When the user has configured all the parameters to their satisfaction they can start miRProf by clicking on the “Start” button on the main dialog, or by selecting the “Start” menu item from the run menu. Once running miRProf can be cancelled at any time by clicking on the “Cancel” button or menu item. Note: cancelling a run may not be instant as execution must reach a safe position in the code before cleanly stopping the run.
After the run has completed the results for all samples is available in the main miRProf dialog as shown in the figure below. The configuration of the results table is highly dependent on the grouping options the user has selected. For example, if the user selected to “group organisms”, then the results table will initially only contain one row, called “all combined”. Otherwise the display shows each organism the sRNAs could be mapped to. By clicking on the organism entry, individual detected miRNAs are revealed. Again, the number of miRNAs displayed here can vary dramatically based on the grouping options selected by the user. Each row containing a detected miRNA will contain 4 columns for each sample: raw count (total number of reads in the sRNA dataset matching this miRNA), weighted count (total number of reads in sRNA dataset matching this miRNA divided by the number of times the matching sRNAs aligned to the genome), normalised count (weighted count divided by total number of reads in this sample multiplied by 1 million), and finally the actual sRNA sequences in this sample that matched the miRNA. Any row within the table can be copied by selecting it and then using “ctrl-c” or by bringing up a context menu and selecting the “Copy to Clipboard” item.
Now that the results are available to the user, miRProf can export two types of file: a results table in .csv format and a list of sRNAs matching known miRNAs (in FASTA format). The results table contains a formatted list of reads that match to known miRNAs. It also contains information about redundant (total) and non-redundant (distinct) sequence counts in the input set before and after every filtering step. The csv file can be loaded into any good spreadsheet program.